CN114386708A

CN114386708A - Method and system for predicting door store passenger flow

Info

Publication number: CN114386708A
Application number: CN202210061981.5A
Authority: CN
Inventors: 卢国鸣
Original assignee: Shanghai Xingrong Information Technology Co ltd
Current assignee: Shanghai Xingrong Information Technology Co ltd
Priority date: 2022-01-19
Filing date: 2022-01-19
Publication date: 2022-04-22

Abstract

The embodiment of the application discloses a method for predicting the passenger flow of a store. The method for predicting the store passenger flow comprises the following steps: acquiring related information of a plurality of stores within a preset range; processing the relevant information and the incidence relation of the plurality of stores based on a passenger flow prediction model, and determining the passenger flow of the target store in the target time period; the passenger flow prediction model comprises a passenger flow embedding layer and a prediction layer; the passenger flow embedding layer determines passenger flow characteristics of a plurality of stores; the prediction layer determines the passenger flow volume of the target store in the target time period based on the flow characteristics of the plurality of stores.

Description

Method and system for predicting door store passenger flow

Technical Field

The present disclosure relates to the field of computer information processing, and in particular, to a method and a system for predicting the amount of store traffic.

Background

The accurate prediction of the passenger flow of the store in a certain time period in the future is of great significance to the operation of the store. For example, store managers may more rationally schedule employee shifts and hours, prepare quantities of merchandise to be sold, guide retailers in their stocking and inventory plans based on predicted passenger flow. However, the passenger flow of the store is influenced by many factors, and it is difficult to accurately predict the passenger flow in a certain time period in the future.

Therefore, it is desirable to provide a method and a system for predicting store passenger flow, so as to obtain a more accurate predicted value of store passenger flow with higher efficiency.

Disclosure of Invention

One embodiment of the present disclosure provides a method for predicting store passenger flow, including: acquiring relevant information of a plurality of stores within a preset range, wherein the plurality of stores comprise a target store and a plurality of associated stores in association with the target store, and the relevant information comprises information of the plurality of stores and passenger flow volume in a historical target time period before a target time point; processing the relevant information of the plurality of stores and the incidence relation based on a passenger flow prediction model, and determining the passenger flow of the target store in a target time period after the target time point; the passenger flow prediction model is a machine learning model and comprises a passenger flow embedding layer and a prediction layer; wherein the passenger volume embedding layer determines passenger volume characteristics of the plurality of stores; the passenger flow embedding layer is a GNN model, the plurality of stores are used as nodes of a graph, and the incidence relation is used as an edge of the graph; the prediction layer determines the passenger flow of the target store in the target time period based on the passenger flow characteristics of the plurality of stores, and the prediction layer is a full connection layer; the passenger flow prediction model is obtained through training in an end-to-end learning mode.

One embodiment of the present specification provides a system for predicting the amount of store traffic, which includes an obtaining module and a determining module; the acquisition module is used for acquiring related information of a plurality of stores within a preset range, wherein the plurality of stores comprise a target store and a plurality of associated stores in association with the target store, and the related information comprises information related to passenger flow of the plurality of stores in a historical target time period before a target time point; the determining module is used for processing the relevant information of the plurality of stores and the incidence relation based on a passenger flow prediction model and determining the passenger flow of the target store in a target time period after the target time point; the passenger flow prediction model is a machine learning model and comprises a passenger flow embedding layer and a prediction layer; wherein the passenger volume embedding layer determines passenger volume characteristics of the plurality of stores; the passenger flow embedding layer is a GNN model, the plurality of stores are used as nodes of a graph, and the incidence relation is used as an edge of the graph; the prediction layer determines the passenger flow of the target store in the target time period based on the flow characteristics of the plurality of stores, and the prediction layer is a full connection layer; the passenger flow prediction model is obtained through training in an end-to-end learning mode.

One of the embodiments of the present specification provides an apparatus for predicting store traffic, including a processor, where the processor is configured to execute the method for predicting store traffic.

One of the embodiments of the present specification provides a computer-readable storage medium, where the storage medium stores computer instructions, and when the computer reads the computer instructions in the storage medium, the computer executes the method for predicting store passenger flow.

Drawings

The present description will be further explained by way of exemplary embodiments, which will be described in detail by way of the accompanying drawings. These embodiments are not intended to be limiting, and in these embodiments like numerals are used to indicate like structures, wherein:

FIG. 1 is a schematic diagram of an application scenario of a store traffic prediction system 100 according to some embodiments of the present description;

FIG. 2 is a flow diagram illustrating an exemplary architecture for predicting passenger flow at a target store for a target time period using a passenger flow prediction model in accordance with some embodiments of the present description;

FIG. 3 is an exemplary schematic diagram of a store flow graph, shown in accordance with some embodiments of the present description;

FIG. 4 is an exemplary flow diagram of a method of training a passenger flow prediction model according to some embodiments described herein;

FIG. 5 is an exemplary flow chart of a method of obtaining current store traffic/historical target time period traffic, according to some embodiments of the present description.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings used in the description of the embodiments will be briefly described below. It is obvious that the drawings in the following description are only examples or embodiments of the present description, and that for a person skilled in the art, the present description can also be applied to other similar scenarios on the basis of these drawings without inventive effort. Unless otherwise apparent from the context, or otherwise indicated, like reference numbers in the figures refer to the same structure or operation.

It should be understood that "system", "apparatus", "unit" and/or "module" as used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, other words may be substituted by other expressions if they accomplish the same purpose.

As used in this specification and the appended claims, the terms "a," "an," "the," and/or "the" are not intended to be inclusive in the singular, but rather are intended to be inclusive in the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that steps and elements are included which are explicitly identified, that the steps and elements do not form an exclusive list, and that a method or apparatus may include other steps or elements.

Flow charts are used in this description to illustrate operations performed by a system according to embodiments of the present description. It should be understood that the preceding or following operations are not necessarily performed in the exact order in which they are performed. Rather, the various steps may be processed in reverse order or simultaneously. Meanwhile, other operations may be added to the processes, or a certain step or several steps of operations may be removed from the processes.

FIG. 1 is a schematic diagram of an application scenario of a store traffic prediction system 100 according to some embodiments of the present description.

Included in the application scenario may be a server 110, a network 120, at least two store terminals 130, a database 140, and other data sources 150. The server 110 may include a processing device 112.

In some embodiments, the store traffic prediction system 100 can implement the methods and/or processes disclosed herein to obtain a predicted value of the traffic for the target store for the target time period after the target point in time.

In some application scenarios, a user using at least two store terminals 130 may include a store owner of the store and may also include service personnel within the store.

In some embodiments, the information about the passenger flow of the plurality of stores may be obtained through at least two store terminals 130 and/or other data sources 150, and the predicted passenger flow value of the target store in the target time period after the target time point is determined after the processing of the server 110. Server 110 may retrieve data on database 140 or save data to database 140 at the time of processing. Operations in this description may be performed by processing device 112 executing program instructions. The above-described method is merely for convenience of understanding, and the present system may also be implemented in other possible operation modes.

In some examples, different functions, such as data filtering, querying, preprocessing, model training, model execution, etc., may be performed on different devices, respectively, without limitation.

In some embodiments, the database 140 may be included in the server 110, the at least two store terminals 130, and possibly other system components.

In some embodiments, the processing device 112 may be included in the server 110, the at least two store terminals 130, and possibly other system components.

The server 110 and the at least two store terminals 130 may be connected via the network 120, and the database 140 may be connected to the server 110 via the network 120, directly connected to the server 110, or internal to the server 110. A database 140, other data sources 150, may be connected to the network 120 to communicate with one or more components of the forecast system 100 of store traffic. One or more components of the store traffic prediction system 100 may access data or instructions stored in the database 140 and other data sources 150 via the network 120.

The server 110 may be used to manage resources and process data and/or information from at least one component of the present system or an external data source (e.g., a cloud data center). In some embodiments, the server 110 may be a single server or a group of servers. The set of servers can be centralized or distributed (e.g., the servers 110 can be a distributed system), can be dedicated, or can be serviced by other devices or systems at the same time. In some embodiments, the server 110 may be regional or remote. In some embodiments, the server 110 may be implemented on a cloud platform, or provided in a virtual manner. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-tiered cloud, and the like, or any combination thereof.

In some embodiments, the server 110 may include a processing device 112. Processing device 112 may process data and/or information obtained from other devices or system components. The processor may execute program instructions based on the data, information, and/or processing results to perform one or more of the functions described herein. In some embodiments, the processing device 112 may include one or more sub-processing devices (e.g., single core processing devices or multi-core processing devices). By way of example only, the processing device 112 may include a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), an Application Specific Instruction Processor (ASIP), a Graphics Processing Unit (GPU), a Physical Processing Unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a programmable logic circuit (PLD), a controller, a micro-controller unit, a Reduced Instruction Set Computer (RISC), a microprocessor, or the like or any combination thereof.

The network 120 may connect the various components of the system and/or connect the system with external resource components. Network 120 enables communication between the various components and with other components outside the system to facilitate the exchange of data and/or information. In some embodiments, the network 120 may be any one or more of a wired network or a wireless network. For example, network 120 may include a cable network, a fiber optic network, a telecommunications network, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Public Switched Telephone Network (PSTN), a bluetooth network, a ZigBee network (ZigBee), Near Field Communication (NFC), an in-device bus, an in-device line, a cable connection, and the like, or any combination thereof. The network connection between the parts can be in one way or in multiple ways. In some embodiments, the network may be a point-to-point, shared, centralized, etc. variety of topologies or a combination of topologies. In some embodiments, network 120 may include one or more network access points. For example, the network 120 may include wired or wireless network access points, such as base stations and/or network switching points 120-1, 120-2, …, through which one or more components of the store traffic prediction system 100 may connect to the network 120 to exchange data and/or information.

Store terminal 130 refers to one or more terminal devices or software within a store. In some embodiments, one or more users may be using the store terminal 130, may include the owner of the store, and may also include other associated personnel. In some embodiments, the store terminal 130 may be one or any combination of a mobile device, a tablet computer, a laptop computer, a desktop computer, or other device having input and/or output capabilities.

Database 140 may be used to store data and/or instructions. Database 140 is implemented in a single central server, multiple servers or multiple personal devices connected by communication links. In some embodiments, database 140 may include mass storage, removable storage, volatile read-write memory (e.g., random access memory RAM), read-only memory (ROM), the like, or any combination thereof. Illustratively, mass storage may include magnetic disks, optical disks, solid state disks, and the like. In some embodiments, database 140 may be implemented on a cloud platform.

Other data sources 150 may be used to provide one or more sources of other information for the system. The other data sources 150 may be one or more devices, may be one or more application program interfaces, may be one or more database query interfaces, may be one or more protocol-based information acquisition interfaces, may be other ways in which information may be acquired, or may be a combination of two or more of the above. The information provided by the information source may be already present when the information is extracted, may be temporarily generated when the information is extracted, or may be a combination of the above. In some embodiments, other data sources 150 may be used to provide the system with store review information, historical target time period order information, weather information, and traffic conditions around the store, among other things.

FIG. 2 is a flow diagram illustrating an exemplary architecture for predicting passenger flow for a target store at a target time period using a passenger flow prediction model in accordance with some embodiments of the present description. As shown in fig. 2, the process 200 includes the following steps. In some embodiments, flow 200 may be performed by a processing device (e.g., processing device 112).

Step 210, obtaining relevant information of a plurality of stores within a preset range, wherein the plurality of stores comprise a target store and a plurality of associated stores having an association relation with the target store. In some embodiments, this step may be performed by the acquisition module.

In some embodiments, the store may be a physical sales service location that delivers goods or services to end consumers or institutions. In some embodiments, the target store may be a store whose traffic volume in the target time period needs to be predicted, and the associated store may be a store that has an association relationship with the target store. For example, in a certain mall, there are a bakery, a coffee shop, a clothing shop, a fast food restaurant, a movie theater, and the like, wherein the fast food restaurant may be a target store, and other stores in the mall may be associated stores. In some embodiments, the preset range may be one set based on predicted need. For example, the predetermined range may be within a certain market. For another example, the preset range may be a certain area of a business area centered around a target store.

In some embodiments, the relevant information may be data and/or information that directly or indirectly reflects the amount of traffic in the store at the target time period. For example, the related information may include evaluated information of stores. In some embodiments, the relevant information may include information relating to the amount of traffic for a plurality of stores within a historical target time period prior to the target point in time. The target time point may be a current or future time point, the target time period may be a time period after the target time point, and the historical target time period may be a time period before the target time point. For example, the current time is 9:00 am of Beijing time, the target time period may be 14:00-15:00 or 14:30-15:30 of Beijing time of the day, and the historical time period may be 10:00-11:00, 11:00-12:00, etc. of Beijing time of the day. In some embodiments, the target time period and the historical time period may be for a particular time period. For example, the target time period and the historical time period may be weekends, wednesdays, or the like. The target time period and the historical time period may be specific time periods of the day. For example, the historical time period may be 15:00-17:00 in the afternoon, and the target time period may be a time period for eating in the evening, e.g., 17:00-18:00, 18:00-19:00, 19:00-20:00, etc.

In some embodiments, the target time period and the historical time period may be for a particular time period. For example, the target time period and the historical time period may be weekends, wednesdays, or the like. The target time period and the historical time period may be specific time periods of the day. For example, the historical time period may be 15:00-17:00 in the afternoon, and the target time period may be a time period for eating in the evening, e.g., 17:00-18:00, 18:00-19:00, 19:00-20:00, etc.

In some embodiments, the relevant information may include at least one of store type, rated information, historical goal period order information, historical goal period traffic, and current traffic.

In some embodiments, store types may be determined by the type of goods and/or services the store operates. For example, store types may include malls, convenience stores, supermarkets, coffee shops, restaurants, libraries, movie theaters, and the like.

In some embodiments, the rated information may be a record of the customer's reviews of the store. For example, the evaluated information may be a comment record related to a store displayed on an online platform such as a comment website, a takeout APP, or the like. The evaluated information may reflect the ordinary amount of traffic in the store. For example, the more information that is evaluated, the more people who are going to a store for consumption. For another example, the evaluated information has a high score, which indicates that more people may go to a store for consumption in the future.

In some embodiments, the historical target time period may be a time period corresponding to the target time period that has elapsed. For example, the current time is 9:00 am of Beijing hours on Tuesday, the target time period is 5:00 PM of Beijing hours on Tuesday to 6:00 PM, and the historical target time period may be 5:00 PM of Beijing hours on Monday to 6:00 PM. A historical goal time period order may refer to an order that was received and/or completed during a historical goal time period. In some embodiments, the historical target time period order information may include the number of orders within the historical target time period, the order completion status (e.g., completed or cancelled, etc.), the rating information of the order (e.g., good or bad, etc.), the type of goods ordered, the number of goods ordered, the amount of goods ordered, etc., or any combination thereof.

In some embodiments, the volume of traffic may be the number of people in the store. In some embodiments, the current passenger volume may be the passenger volume at the current point in time. In some embodiments, the historical target time period passenger volume may be the passenger volume within the historical target time period. The current passenger flow volume and the historical target time period passenger flow volume can reflect the passenger flow volume of the store in a certain time period in the future to a greater extent. For example, if the target time period is the peak meal time of saturday and the peak meal time of saturday is taken as the historical target time period, the passenger flow in the historical target time period may reflect the passenger flow in the target time period to a greater extent.

The processing device may obtain store-related information in a variety of ways. For example, the processing device may obtain the type of store through registration information of the store obtained by the other data source 150. For another example, the processing device may acquire the evaluated information of the store, the historical target time period order information, and the like by querying the relevant information of the online platform.

One embodiment of a processing device to obtain current passenger flow and historical target time period passenger flow may be found in fig. 5 and its associated description.

In some embodiments, the related information further includes wireless signal related information within the store. The wireless signal related information may include, but is not limited to: whether the store provides wireless signals, the way of accessing the wireless signals in the store, the number of the signal intensity in a certain range of the store larger than a threshold value, and the like. The wireless signal related information may affect the amount of traffic in the store to some extent. For example, under the same conditions, a consumer may stay longer in a store that provides wireless signals because the wireless signals in the store may be used for entertainment or work. As another example, for a consumer who needs to use a network in a store (e.g., a coffee shop), the store with a stronger wireless signal is preferred.

In some embodiments, the associative relationship may include a variety of relationships between stores (e.g., between the target store and the associated store, between the associated store and the associated store) that are related to the amount of traffic. For example, the relevance of the business between the target store and the associated store.

In some embodiments, the associative relationships include distance relationships, location relationships, and/or business relationships between multiple stores.

In some embodiments, the distance relationship may represent how far and near between stores. The distance relationship may reflect to some extent the interplay of the amount of traffic between stores. For example, as the volume of play on a casino increases, the volume of play in the toy stores adjacent to it will increase accordingly.

In some embodiments, the location relationship includes at least one of whether the target store and the associated store are in the same building, across a street, or belong to the same business establishment. The positional relationship may reflect the mutual influence of the passenger flow volume between stores to some extent. For example, an associated store in the same building as the target store may have a greater impact on the amount of traffic in the target store than an associated store that needs to cross the street.

In some embodiments, business relationships include business similarities and/or business inheritances.

In some embodiments, business similarity may be the degree of similarity of business between stores. For example, business similarity between stores selling mobile phone accessories is high, and business similarity between stores selling mobile phone accessories and restaurant stores is low. Business similarity may reflect to some extent the interplay of passenger traffic between stores. For example, if the target store is a restaurant, the influence of the passenger flow volume on the passenger flow volume of the target store is large when the target store is a store associated with the restaurant.

In some embodiments, business inheritance may be the likelihood of a consumer going to one store for consumption, while going to another store for consumption, or subsequently. For example, because a large percentage of consumers may choose to eat lunch next to a movie theater after watching the movie, there is a high level of inheritance between the movie theater and restaurants near the movie theater.

Step 220, processing the relevant information and the incidence relation of the plurality of stores based on the passenger flow prediction model, and determining the passenger flow of the target store in the target time period after the target time point. In some embodiments, this step may be performed by the passenger flow prediction module.

In some embodiments, the traffic prediction model may be a trained machine learning model for predicting store traffic for a certain period of time in the future. The passenger flow prediction model can be constructed based on various machine learning models, such as a Long Short-Term Memory network (LSTM) model, a Gate-controlled round-robin Unit (GRU) model, a Graph Neural Network (GNN) model, and the like.

In some embodiments, the passenger flow prediction model may include a passenger flow embedding layer and a prediction layer.

In some embodiments, the passenger flow embedding layer may be configured to process the related information and the association relationship to extract passenger flow characteristics of the plurality of stores. In some embodiments, the input to the passenger volume embedding layer may be feature data obtained by preprocessing (e.g., normalizing, discretizing, etc.) the relevant information and the association, and the output of the passenger volume embedding layer may be passenger volume features of the plurality of stores in vector form. The passenger volume embedding layer can be constructed based on various machine learning models. In some embodiments, the passenger volume embedding layer may include, but is not limited to: long Short-Term Memory network (LSTM) models, gated round-robin Unit (GRU) models, Graph Neural Networks (GNN) models, and the like.

In some embodiments, the passenger volume embedding layer is a GNN model, and further description of the GNN model can be found in fig. 3 and its associated description.

In some embodiments, the prediction layer may determine the traffic volume of the target store for a target time period after the target point in time based on the traffic volume characteristics of the target store extracted by the traffic volume embedding layer. In some embodiments, the input to the prediction layer may be a traffic characteristic of the target store and the output of the prediction layer may be a traffic of the target store for a target time period after the target point in time. The amount of traffic output by the prediction layer may be represented in a variety of ways. For example, a passenger flow level may be used to represent passenger flow. The traffic level may be represented by a number, for example, 1 indicates a traffic volume of less than 3 people, and 10 indicates a situation where the number of people in the store has reached saturation. The prediction layer may be constructed based on a variety of machine learning models. In some embodiments, the prediction layer may include, but is not limited to: decision trees (Decision trees), Random Forest (RF), Support Vector Machines (SVMs), and the like.

In some embodiments, the prediction layer may be a fully connected layer. The full-link layer may predict the passenger volume of the target store in the target time period after the target time point based on the passenger volume characteristics of the target store extracted by the passenger volume embedding layer.

In some embodiments, the passenger flow prediction model is obtained by training in an end-to-end learning manner, and for further description, refer to fig. 4 and its related description.

In the method described in some embodiments of the present specification, when predicting the passenger flow volume of the target store, not only the relevant information of the target store itself is considered, but also the mutual influence of the passenger flow volumes among the stores is considered, so that the prediction result is more accurate. In some embodiments, the machine learning model is used for processing the related information of a plurality of stores and the association relationship among the plurality of stores, so that more accurate test results can be obtained with higher processing efficiency.

It should be noted that the above description related to the flow 200 is only for illustration and description, and does not limit the applicable scope of the present specification. Various modifications and alterations to flow 200 will be apparent to those skilled in the art in light of this description. However, such modifications and variations are intended to be within the scope of the present description. For example, the input of the prediction layer of the passenger flow prediction model may be passenger flow characteristics of a plurality of stores, and the output may be passenger flow of the plurality of stores in the target time period.

FIG. 3 is an exemplary schematic diagram of a store flow graph, shown in accordance with some embodiments herein.

In some embodiments, the passenger volume embedding layer in the passenger volume prediction model is a GNN model with multiple stores as nodes of the graph and associations as edges of the graph. For ease of description, in one or more embodiments of the present description, the graph corresponding to the GNN model is referred to as a store-front traffic graph.

As shown in fig. 3, the store traffic map includes a plurality of nodes (e.g., store 1, store 2, etc.), where store 1 is the target store and the other stores are associated stores. The store customer flow graph also includes a plurality of edges (e.g., edge 1, edge 2, etc.), wherein the edges are used to represent associations between the plurality of stores.

In some embodiments, the node characteristics of the gate store passenger flow graph may include relevant information and environmental information. In some embodiments, the environmental information may include, but is not limited to: weather information, traffic conditions around stores, etc. The environmental information may affect the amount of passenger traffic in the store to some extent. For example, in rainy weather versus sunny days, store traffic may be relatively low. For example, in the case of traffic congestion around a store, the customer may select another store with convenient traffic to consume, and thus the customer flow of the store is also affected. For more description of the related information, reference may be made to fig. 2 and its associated description.

In some embodiments, the processing device may preprocess the relevant information and the environmental information as node features of the storefront traffic graph. For example, the processor may digitize store types, information being evaluated, and the like using one-hot encoding. As another example, the processor may bin the order quantity to discretize it.

In some embodiments, edges of the storefront traffic graph may be characterized by associations. For more description of the association relationship, refer to fig. 2 and its related description.

In some embodiments, the processing device may preprocess the associations as an edge feature of a storefront traffic graph. For example, the processor may digitize location relationships, business relationships, and the like using one-hot encoding. For another example, the processor may bin the distances between stores to discretize them.

According to the method described in some embodiments of the present specification, the GNN model is used to process the relevant information of the target store and the relevant information of the associated store, so that the passenger flow characteristics extracted by the passenger flow embedding layer fully fuse the influence of the passenger flow of the associated store on the passenger flow of the target store, and the passenger flow prediction result of the target store is more accurate.

FIG. 4 is an exemplary flow diagram of a method of training a passenger flow prediction model according to some embodiments described herein.

At step 410, training samples are obtained. In some embodiments, this step may be performed by a training module.

In some embodiments, the training sample may contain sample-related information and sample environment information for a plurality of sample stores, including a sample target store and a sample association store, and a sample association relationship between the plurality of sample stores. The label of the training sample is the passenger flow volume value of the sample target store in the sample historical target time period. The training samples may be historical real data, and the processing device may obtain the training samples in a common manner, such as network downloading, database extraction, and the like.

And step 420, performing end-to-end training on the passenger flow prediction model based on the training samples. In some embodiments, this step may be performed by a training module.

In some embodiments, the passenger flow prediction model may be trained end-to-end based on training samples. In the end-to-end training process of the passenger flow prediction model, the processing device can input training sample data into a passenger flow embedding layer of the passenger flow prediction model, and then obtains an error value by using a loss function based on a passenger flow predicted value output by a prediction layer of the passenger flow prediction model, wherein the loss function is constructed based on a predicted value and a label of the sample data. Then, the processing device may calculate gradients corresponding to all parameters to be learned in the passenger flow prediction model (including parameters of the passenger flow embedding layer, parameters of the prediction layer, and the like) by using a back propagation algorithm (BP algorithm) according to the error value, and adjust the parameters of the passenger flow prediction model (including parameters of the passenger flow embedding layer, parameters of the prediction layer, and the like) by combining a random gradient descent algorithm until the model converges.

In the method described in some embodiments of the present specification, in the training process of the passenger flow prediction model, a data tagging process of passenger flow characteristics output by the passenger flow embedding layer is omitted through end-to-end learning, so that a large amount of workload of data tagging is saved, and the time period of model training is reduced. Meanwhile, the end-to-end training method can fully consider the correlation among the parameters of each processing layer, so that the passenger flow prediction model obtained through end-to-end learning can more accurately predict the passenger flow of the store.

It should be noted that the above description related to the flow 400 is only for illustration and description, and does not limit the applicable scope of the present specification. Various modifications and changes to flow 400 will be apparent to those skilled in the art in light of this description. However, such modifications and variations are intended to be within the scope of the present description. For example, step 420 may be split into two steps 420-1 and 420-2, a loss function is constructed in step 420-1, and the model is trained in step 420-2.

At step 510, at least one frame of image data is extracted from the video data of the stores in the current time period/the historical target time period. In some embodiments, this step may be performed by the passenger flow volume acquisition module.

In some embodiments, the processing device may obtain, through the store terminal 130, video data of the store at the current time period and/or the historical target time period, wherein the video data may be captured by the monitoring device. The processing device may also obtain the video data in other manners, which is not limited herein. The format of the video data may include, but is not limited to: MP4 format, MKV format, M4A format, MP3 format, FLV format, and 3GP format, etc.

In some embodiments, the processing device may extract at least one frame of image data from the video data for subsequent processing. For example, the processing device may extract one or more pieces of image data with higher resolution from the video data.

In some embodiments, in order to obtain image data with higher picture quality, the processing device may extract image data with a code stream value greater than a preset threshold from the video data. The preset threshold may be a code stream value that is empirically preset.

And step 520, processing at least one frame of image data based on the trained passenger flow volume judgment model, and determining the current passenger flow volume/the passenger flow volume of the store in the historical target time period. In some embodiments, this step may be performed by the passenger flow volume acquisition module.

In some embodiments, the passenger flow volume assessment model may process the image data to obtain current passenger flow volumes and/or historical target time period passenger flow volumes for the store. For example, the current passenger flow volume of the store can be obtained by inputting image data corresponding to the current time zone into the passenger flow volume determination model. For another example, the passenger flow volume in the past target time zone of the store can be obtained by inputting image data corresponding to the past target time zone into the passenger flow volume determination model.

In some embodiments, the input to the passenger flow determination model may be image data. In some embodiments, the output of the passenger flow determination model may be passenger flow. The passenger flow output by the passenger flow judgment model can have various expression modes. For example, a passenger flow level may be used to represent passenger flow. Also for example, the number of people may be used to represent the volume of traffic.

In some embodiments, the passenger flow judgment model may include, but is not limited to: convolutional neural network models, deep neural network models, and the like.

In some embodiments, the passenger flow volume judgment model is a convolutional neural network model, and the parameter of the convolutional kernel of the passenger flow volume judgment model is obtained based on the migration of the convolutional kernel in the pre-trained image recognition model. For example, the image recognition model may be trained using multiple sets of sample data, where the sample data includes image data and labels as model inputs, the labels are the number of people in the image, and the convolution kernel of the trained image recognition model is used as the convolution kernel of the passenger flow volume determination model. In some embodiments, the convolution kernel in the image recognition model may include multiple network layers, and the following operations may be performed on the input image data: convolution operations (Convolution, Conv), Normalization operations (Normalization), activation operations (e.g., ReLU functions), and dimension reduction operations (e.g., max pooling).

In the method described in some embodiments of the present specification, the convolutional neural network model is used to process image data, and thus more accurate current passenger flow volume/historical target time period passenger flow volume can be obtained.

It should be noted that the above description related to the flow 500 is only for illustration and description, and does not limit the applicable scope of the present specification. Various modifications and changes to flow 500 may occur to those skilled in the art, given the benefit of this description. However, such modifications and variations are intended to be within the scope of the present description. For example, the passenger flow judgment model may be constructed based on a deep neural network model.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is to be regarded as illustrative only and not as limiting the present specification. Various modifications, improvements and adaptations to the present description may occur to those skilled in the art, although not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present specification and thus fall within the spirit and scope of the exemplary embodiments of the present specification.

Also, the description uses specific words to describe embodiments of the description. Reference throughout this specification to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the specification is included. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, some features, structures, or characteristics of one or more embodiments of the specification may be combined as appropriate.

Additionally, the order in which the elements and sequences of the process are recited in the specification, the use of alphanumeric characters, or other designations, is not intended to limit the order in which the processes and methods of the specification occur, unless otherwise specified in the claims. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the present specification, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the embodiments. This method of disclosure, however, is not intended to imply that more features than are expressly recited in a claim. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Numerals describing the number of components, attributes, etc. are used in some embodiments, it being understood that such numerals used in the description of the embodiments are modified in some instances by the use of the modifier "about", "approximately" or "substantially". Unless otherwise indicated, "about", "approximately" or "substantially" indicates that the number allows a variation of ± 20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximations that may vary depending upon the desired properties of the individual embodiments. In some embodiments, the numerical parameter should take into account the specified significant digits and employ a general digit preserving approach. Notwithstanding that the numerical ranges and parameters setting forth the broad scope of the range are approximations, in the specific examples, such numerical values are set forth as precisely as possible within the scope of the application.

For each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, etc., cited in this specification, the entire contents of each are hereby incorporated by reference into this specification. Except where the application history document does not conform to or conflict with the contents of the present specification, it is to be understood that the application history document, as used herein in the present specification or appended claims, is intended to define the broadest scope of the present specification (whether presently or later in the specification) rather than the broadest scope of the present specification. It is to be understood that the descriptions, definitions and/or uses of terms in the accompanying materials of this specification shall control if they are inconsistent or contrary to the descriptions and/or uses of terms in this specification.

Finally, it should be understood that the embodiments described herein are merely illustrative of the principles of the embodiments of the present disclosure. Other variations are also possible within the scope of the present description. Thus, by way of example, and not limitation, alternative configurations of the embodiments of the specification can be considered consistent with the teachings of the specification. Accordingly, the embodiments of the present description are not limited to only those embodiments explicitly described and depicted herein.

Claims

1. A method for predicting store traffic, the method comprising:

acquiring relevant information of a plurality of stores within a preset range, wherein the plurality of stores comprise a target store and a plurality of associated stores in association with the target store, and the relevant information comprises information of the plurality of stores and passenger flow volume in a historical target time period before a target time point;

processing the relevant information of the plurality of stores and the incidence relation based on a passenger flow prediction model, and determining the passenger flow of the target store in a target time period after the target time point;

the passenger flow prediction model is a machine learning model and comprises a passenger flow embedding layer and a prediction layer; wherein,

the passenger flow volume embedding layer determines passenger flow volume characteristics of the plurality of stores; the passenger flow embedding layer is a GNN model, the plurality of stores are used as nodes of a graph, and the incidence relation is used as an edge of the graph;

the prediction layer determines the passenger flow of the target store in the target time period based on the passenger flow characteristics of the plurality of stores, and the prediction layer is a full connection layer;

the passenger flow prediction model is obtained through training in an end-to-end learning mode.

2. The method of claim 1, wherein the related information further comprises current passenger volume and/or the historical target time period passenger volume, wherein the current passenger volume or the historical target time period passenger volume is obtained by:

for each of the plurality of stores,

extracting at least one frame of image data from the video data of the store in the current time period or the historical target time period, wherein the at least one frame of image data is an image of which the code stream value is greater than a preset threshold value in the video data;

and processing the at least one image data based on the trained passenger flow volume judgment model to determine the current passenger flow volume of the store or the passenger flow volume of the historical target time period.

3. The method of claim 2, wherein the passenger flow volume judgment model is a convolutional neural network model, and parameters of a convolutional kernel of the passenger flow volume judgment model are obtained based on a convolutional kernel migration in a pre-trained image recognition model.

4. The method of claim 2, wherein the related information further comprises in-store wireless signal related information.

5. A prediction system of door shop passenger flow is characterized by comprising an acquisition module and a determination module;

the acquisition module is used for acquiring related information of a plurality of stores within a preset range, wherein the plurality of stores comprise a target store and a plurality of associated stores in association with the target store, and the related information comprises information related to passenger flow of the plurality of stores in a historical target time period before a target time point;

the determining module is used for processing the relevant information of the plurality of stores and the incidence relation based on a passenger flow prediction model and determining the passenger flow of the target store in a target time period after the target time point;

6. The system of claim 5, wherein the relevant information further comprises current passenger volume and/or the historical target time period passenger volume, the current passenger volume or the historical target time period passenger volume being obtained by:

for each of the plurality of stores,

7. The system of claim 5, wherein the passenger flow volume judgment model is a convolutional neural network model, and parameters of a convolutional kernel of the passenger flow volume judgment model are obtained based on a convolutional kernel migration in a pre-trained image recognition model.

8. The system of claim 6, wherein the related information further comprises in-store wireless signal related information.

9. A prediction device of store passenger flow, comprising a processor, wherein the processor is used for executing the prediction method of store passenger flow according to any one of claims 1 to 4.

10. A computer-readable storage medium storing computer instructions which, when read by a computer, cause the computer to perform the method of predicting store occupancy according to any one of claims 1 to 4.